Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

Detection and Monitoring of Microparticles under

Skin by Optical Coherence Tomography as an

Approach to Continuous Glucose Sensing using

Implanted Retroreflectors

Shang Wang, Tim Sherlock, Betsy Salazar, Narendran Sudheendran, Ravi Kiran Manapuram, Katerina Kourentzi,

Paul Ruchhoeft, Member, IEEE, Richard C. Willson, and Kirill V. Larin, Member, IEEE

Abstract—We demonstrate the feasibility of using optical

coherence tomography (OCT) to image and detect 2.8 m

diameter microparticles (stationary and moving) on a highly-

reflective gold surface both in clear media and under skin in vitro.

The OCT intensity signal can clearly report the microparticle

count, and the OCT response to the number of microparticles

shows a good linearity. The detectability of the intensity change

(2.9%±0.5%) caused by an individual microparticle indicates the

high sensitivity of monitoring multiple particles using OCT. An

optical sensing method based on this feasibility study is described

for continuously measuring blood sugar levels in the

subcutaneous tissue, and a molecular recognition unit is designed

using competitive binding to modulate the number of bound

microparticles as a function of glucose concentration. With

further development, an ultra-small, implantable sensor might

provide high specificity and sensitivity for long-term continuous

monitoring of blood glucose concentration.

Index Terms—Biomedical monitoring, magnetic microparticles,

optical coherence tomography, skin, retroreflection

I. INTRODUCTION

Frequent blood glucose monitoring is highly recommended

for patients suffering from diabetes in order to prevent acute

complications and reduce the risk of long-term problems [1].

However, the most widely used current sensing method

requires collection of a droplet of blood from a fingertip

through a puncture, which is painful and uncomfortable and

acts as a deterrent to frequent blood glucose measurements. To

improve management of diabetes and make blood glucose

testing easier and less painful, an implantable (or noninvasive)

sensor, capable of providing continuous measurement of blood

glucose levels, offers the opportunity for improved therapeutic

interventions [2, 3].

Both electrochemical and optical methods have been

extensively studied as possible approaches to an implantable

(noninvasive) glucose sensor. Owing to concerns for potential

leaching of mediators, the majority of implantable sensors

applying electrochemical methods are based on the

irreversible oxidation of glucose [4]. Most electrochemical

sensors have issues of biocompatibility, short storage and

operational lifetimes, and response drift [4, 5]. Optical

methods based on the measurement of photons interacting

with tissues have certain advantages, such as being free of a

reference electrode and consumable reagents, and being

capable of sterile remote sensing and multiplexing [5, 6].

Optical glucose sensing methods can be divided into two

broad categories: one approach is to measure the direct effects

of glucose molecules on the properties of light (e.g. intensity

and polarization), and the other is to use a molecular

recognition technique to transduce the glucose concentration

into a detectable light signal. Both of these methods have the

potential to be developed as a noninvasive or implantable

optical sensor.

Several optical techniques have been proposed for

noninvasive blood glucose sensing [6-8]. Near-infrared (NIR)

spectroscopy provides information on glucose concentration

by measuring the transmission or reflectance of light. As

glucose can alter optical absorption properties of tissues,

changes in the transmission or reflectance spectrum can be

linked to the changes in blood glucose concentration [9, 10].

However, the tissue absorption coefficient also depends on

factors such as temperature and changes in the concentration

of other analytes, which highly affects the accuracy of sensing.

Raman spectroscopy derives an estimation of glucose

concentration through analysing the scattered light influenced

by the oscillation and rotation of glucose molecules [11, 12].

Although the Raman spectrum of glucose is well differentiable

from that of other molecules, the long spectral acquisition time

and the presence of many other unrelated compounds can

S. Wang, B. Salazar, N. Sudheendran, and K. V. Larin* are with the

Department of Biomedical Engineering, University of Houston, Houston, TX

77204, USA (e-mail: wangshang1218@hotmail.com; bhsalazar@uh.edu;

nsudheendran@uh.edu; klarin@uh.edu). * Correspondence: klarin@uh.edu.

T. Sherlock and P. Ruchhoeft are with the Department of Electrical and

Computer Engineering, University of Houston, Houston, TX 77204, USA (e-

mail: the.tim.sherlock@gmail.com; pruchhoeft@uh.edu).

R. K. Manapuram was with the Department of Mechanical Engineering,

University of Houston, Houston, TX 77204, USA, and is now with Bioptigen

Inc., Morrisville, NC 27560, USA (e-mail: kiransgis@yahoo.com).

K. Kourentzi and R. C. Willson are with the Department of Chemical and

Biomolecular Engineering, University of Houston, TX 77204, USA (e-mail:

edkourentzi@uh.edu; willson@uh.edu).

R. C. Willson is also affiliated in The Methodist Hospital Research

Institute, Houston, TX 77031, USA.

K. V. Larin is also affiliated at the Department of Molecular Physiology

and Biophysics, Baylor College of Medicine, Houston, TX 77030, USA.

Copyright (c) 2013 IEEE. Personal use of this material is permitted.

However, permission to use this material for any other purposes must be

obtained from the IEEE by sending a request to pubs-permissions@ieee.org.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

strongly limit the clinical application of this method.

Polarimetry utilizes the optical rotatory dispersion of glucose

molecules and reflects the blood glucose concentration by

measuring the change of the polarization plane of linearly

polarized light [13, 14]. An anatomic location suggested to be

well suited for the polarimetric measurement is the anterior

chamber of the eye, as the scattering of light in the eye is

generally very small compared with other types of tissues.

Problems of this method mainly exist in the interference

related to the optical rotation effects from the cornea and other

optically active components in the aqueous humour. Optical

coherence tomography (OCT) allows depth-resolved

assessment of tissue scattering coefficient (which is altered by

the changes of glucose concentration) using a low coherence

interferometer [15-18]. But, similar to NIR spectroscopy,

varying physiological and environmental conditions could also

alter the scattering properties of tissue, which highly affects

the measurement accuracy.

To improve specificity, a major effort has been focused on

designing and implementing a glucose recognition and

reporting unit which can be implanted under the skin and

provide a detectable light property change caused by the

variation of glucose concentration. Methods including kinetic

enzymatic assays using glucose oxidation [19, 20], chemical

binding of glucose to synthetic boronic acid [21, 22], affinity

binding to Concanavalin A (Con A) [23-25], and affinity

binding to glucose binding proteins [26, 27] have been studied

and combined with fluorescence, scattering and fiber optic

techniques for blood glucose sensing. Glucose recognition

utilizing affinity binding to Con A offers certain advantages,

including full reversibility, feasibility of calibration, weak pH-

dependence, and no effect of varying oxygen partial pressure

[5]. Con A is a plant lectin which has four binding sites for

glucose and carbohydrates (e.g. dextran). Since it was first

explored by Schultz et al. using fluorescence detection [23,

24], the application of this method has been extensively

studied with fluorescence labelling and imaging techniques to

improve the glucose sensing ability with high specificity [25,

28-31]. However, nonspecific changes in fluorescence reduce

the glucose sensing accuracy and reliability.

Recently, Ballerstadt et al. investigated the feasibility of

using OCT to monitor the turbidity of an implantable glucose

recognition unit in order to analyse the glucose concentration

[32]. The turbidity change is induced by glucose binding to

Con A. The glucose-specific scattering sensitivity can reach

up to 7% mM-1 when measured in air, and remains 0.26%

mM-1 when tested in a tissue phantom. However, this sensing

method may not be able to maintain sensing specificity and

accuracy when implemented below skin due to the large

attenuation of the OCT signals from tissue. Also, the

aggregation of free Con A will shorten the sensor’s potential

lifetime. Thus, for the purpose of developing highly-specified

implantable glucose sensing method with enhanced sensitivity

and long operation time, we propose a glucose recognition

unit with microparticles-labelled dextran and a highly-

reflective gold mirror coated with Con A. The gold mirror

surface is used to provide a strong reflection signal for

maintaining sufficient sensing accuracy after implantation (for

the final construction, we intend to utilize a micro-

retroreflector-based gold platform [33], which is able to

provide high contrast for the detection of reflected light within

tissue). The immobilization of Con A on the gold mirror

surface greatly reduces its aggregation rate and therefore

extends the effective sensing lifetime. For housing the unit, a

semi-permeable membrane can be applied to allow free

diffusion of glucose molecules while being impermeable to

cells and microparticles. A schematic of the proposed structure

is shown in Figure 1. Glucose recognition is realized by

competitive displacement of glucose and microparticles-

labelled dextran to Con A. With different glucose

concentration, the binding competition of glucose and dextran

to Con A will modulate the microparticle density within the

imaging area on the gold mirror surface. Due to the reversible

property of competitive binding [34], specifically, lower

glucose concentration will enable more microparticles to be

localized on the surface, and increasing glucose concentration

will reduce the amount of microparticles. Therefore, through

monitoring the number of microparticles within the target area,

glucose concentration can be determined. For continuous

glucose monitoring, the sensing unit is expected to be

implanted in the upper layer of the dermis, located

approximately 150-300 µm from the skin surface (illustrated

in Figure 1).

More recently, we have demonstrated that OCT has the

capability of depth-resolved imaging of highly-reflective

micro-structures under tissue up to 910 m in depth [35], and,

thus, OCT could be used to detect the change of the reflection

of light from a gold mirror surface under tissue and to monitor

Fig. 1. Illustration of the sensing element construction and implantation position. The picture of skin anatomy on the right is from National Cancer Institute (for

free reuse), http://visualsonline.cancer.gov/details.cfm?imageid=4604.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

the microparticle number based on the scattering properties of

these particles. Since we expect that the glucose concentration

can be well represented by the amount of bound microparticles,

the sensitivity and specificity of the proposed sensing system

are highly dependent on how well the microparticles can be

resolved by OCT and how well the signal from OCT and the

microparticle density are correlated. In this paper, we

demonstrate results from a feasibility study where we focused

on the quantification of the reflection variation caused by the

change of microparticle number on a mirrored surface and also

the monitoring of the microparticles both in clear media and

under skin tissue using OCT.

II. MATERIALS AND METHODS

A. Microparticles

For the proposed sensing method, the microparticles should

be small in size, easy to manipulate, and able to be coated with

dextran. Thus, our primary selection focused on Dynabeads®

with carboxylic acid groups (Life Technologies Corporation).

Two available sizes with 1.05 m and 2.8 m diameter are

suitable for the construction of the sensing unit. Also, their

superparamagnetic property allows the use of magnets to

control the microparticles’ position within the capsule. It is

well known that the attenuation of the light is dependent,

among others, on the diameter of the scatters and the

wavelength of the light [36]. Typically OCT employs the

central wavelengths around 800 nm and 1300 nm [37]. Thus,

we performed a preliminary study of light attenuation versus

microparticle concentrations using two OCT systems with the

laser sources around these wavelengths. Fig. 2 shows the

dependence of the attenuation coefficient on microparticle

concentrations for the 1.05 m and the 2.8 m diameter

microparticles using OCT system centered at 840 nm and

1325 nm. A linear fit was applied to the data to quantify the

slope of the change, and it can be seen that, in Fig. 2(c), the

change of attenuation coefficient appears to have the highest

slope. Moreover, for the low concentration range ( 200

particles/nL), the slope in Fig. 2(c) is even steeper. This result

indicates the sensitivity of the attenuation coefficient to the

changes in microparticle number is greatest for microparticles

with 2.8 m in diameter at a central wavelength of 840 nm.

Thus, based on this test, Dynabeads® M-270 Carboxylic Acid

(2.8 m diameter) was chosen for the feasibility study of the

proposed sensing method.

B. OCT System and en face OCT Imaging

A spectral domain OCT system with the central wavelength

of 840 nm was chosen for use in this study based on the

preliminary data described in Fig. 2. The schematic of the

system is shown in Fig. 3. The laser source (superluminescent

diode) has a bandwidth of ~49 nm with the output power of

around 20 mW. The OCT system is based on a Michelson

interferometer where the light from the reference and the

sample arms interference. A home-built spectrometer is

utilized to resolve the wavelengths and detect the interference

fringes. In the spectrometer, the collimated beam falls onto a

1200 l/mm holographic transmission grating, and the beam

with spatially-separated wavelengths is focused onto the 2048

pixels of a CCD. The data acquisition from the CCD is

synchronized with the Galvo mirrors in the sample arm that

are used for the two-dimensional transverse scan of the sample.

The A-line (depth-resolved imaging) speed of the OCT system

can reach up to 29 kHz. An axial resolution of ~12 m (in air)

and a transverse resolution of ~8 m can be achieved. The

maximum imaging depth in scattering tissues (assuming

refractive index of 1.4) is around 3 mm.

The position of the gold mirror surface can be located and

selected from the depth-resolved OCT images and the

intensity of the OCT signal at this particular depth represents

the reflection from the gold mirror surface. Because the light

scattering from the microparticles reduces the intensity of the

localized reflected light from the gold mirror surface, an en

face OCT image with the information of reflection shows the

microparticle distribution across the gold mirror surface.

Fig. 3. Schematic of the spectral domain OCT system used for microparticles

monitoring; L–laser, FOC–fiber optic coupler, FA–fiber adaptor, AL–

achromatic lens, TG–transmission grating, C–collimator, A–aperture, M–

mirror, GM-X/Y–Galvo mirror-X/Y, SL–scan lens, S–sample, LS-2D–linear

stage-two-dimension.

Fig. 2. Attenuation coefficient of (a) 1.05 m diameter microparticles

measured with 840 nm central wavelength OCT system, (b) 1.05 m diameter

microparticles measured with 1325 nm central wavelength OCT system, (c)

2.8 m diameter microparticles measured with 840 nm central wavelength

OCT system, and (d) 2.8 m diameter microparticles measured with 1325 nm

central wavelength OCT system. N=50 for the times of measurements

presented in (a), (b), (c) and (d).

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

C. Gold Mirror

To create gold mirror surfaces, 100 mm diameter silicon

wafers were coated with a 10 nm thick layer of titanium

followed by a 100 nm thick layer of gold deposited using

thermal evaporation.

D. Mouse Ear

Skin from a mouse ear was used in the in vitro studies. A

mouse ear was dissected and kept in 0.9% saline before

experiments. During experiments, distilled water was added to

avoid dehydration. Based on the measurement from OCT

three-dimensional images, the skin tissue from the mouse ear

had a thickness of 150-300 m.

E. Experimental Procedure

The experiments for the feasibility study of the proposed

glucose sensing method were performed in three phases.

1) Assessing the variation of intensity from OCT signal

with the change of microparticle number on the gold mirror

surface. Low-concentration (1103 particles/L – 1105

particles/L) microparticle solutions were prepared and

deposited on the surface of gold mirrors. The mirrors were

marked with cross-shaped scratches for the purpose to provide

reference positions. After the microparticles were allowed to

dry on the gold mirror, an optical microscope and the OCT

system were used to image the same regions. The mapping of

the microparticles on the gold mirror surface using en face

OCT images was examined by the corresponding microscope

images. Based on the mapping results, the change of intensity

from OCT signal at the microparticle locations was quantified.

2) Imaging and detecting the movement of microparticles in

clear media. A low-concentration microparticle solution (1

mL, 200 particles/L) was prepared and deposited on the gold

mirror surface of ~172 mm2 area. The microparticles settled

on the mirror surface before the OCT recording was started.

The OCT imaging area was set to be 935187 m2 with a

scanning interval of ~2.3 m. The imaging speed was 2

seconds per two-dimensional scan. A magnetic field was

applied at the time around 3 minutes after the recording

started. The average of the intensity of OCT signal from the

gold mirror surface in each two-dimensional scan was

obtained and recorded over time. Also, en face OCT images at

different time points were generated.

3) Detecting and monitoring the movement of

microparticles under skin tissue in vitro. The skin tissue from

mouse ear was placed on the gold mirror, and a microparticle

solution of 200 particles/L was added between the tissue and

the gold mirror surface, forming a layer with the thickness of

about 0.7–1.0 mm. The experimental setup is shown with a

picture in Fig 4(a). An area of 935187 m2 was scanned by

the OCT system, and a typical three-dimensional OCT image,

including the gold mirror surface and the skin tissue, is shown

in Fig. 4(b). The OCT transverse scanning interval was also

kept at ~2.3 m and the recording time for each two-

dimensional scan was 2 seconds. Similar to the second phase,

the average of the intensity of OCT signal from the gold

mirror surface in each two-dimensional scan was plotted over

time for both before and after applying the magnetic field.

III. RESULTS AND DISCUSSIONS

Selected typical en face OCT images of microparticles

deposited on the gold mirror surface are shown in Fig. 5(a)

and (c). The corresponding optical microscope images of the

same imaging regions are shown in Fig. 5(b) and (d). The

locations of the cross-shaped scratches offered the references

for the mapping of the microparticles. Fig. 5 shows that OCT

is able to correctly map the distribution of microparticles on

the gold mirror surface. The clusters of three microparticles

and the corresponding dark points in the en face OCT images

are called out with the black arrows in Fig. 5.

By mapping the intensity signals from the OCT scan to the

actual microparticle count, the fractional decrease in the

intensity of OCT signal was calculated for single

microparticles and aggregates containing a variety of

microparticle counts. The result is shown in Fig. 6. For single

microparticles and aggregates with 2-8 microparticles, N=4

(samples for measurement). Since the aggregates with more

Fig. 4. (a) Experimental setup for studying the OCT monitoring of

microparticles’ movement under mouse ear skin tissue in vitro. (b) Three-

dimensional OCT image showing the mouse ear skin tissue and the gold

mirror surface with a vertical distance of 0.7-1.0 mm between them.

Fig. 5. (a) and (c) are en face OCT images of microparticles deposited on the

gold mirror surface; (b) and (d) are the corresponding microscopic images of

microparticles deposited on the gold mirror surface. The scale bars are 100

m long and the black arrows point to the dark points, which correspond to

the clusters of three microparticles. Partially magnified pictures from (b) and

(d) are squared with yellow color.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

microparticles are rare, for aggregates with 9, 10 and 11

microparticles, N=1 (sample for measurement). Fig. 6 shows

that when the number of microparticles increases, the intensity

decreases with an approximately linear (R=0.89) relationship.

The intensity change caused by one single microparticle is

2.9%±0.5%. These results indicate that OCT has the ability to

assess the microparticle number through detecting the change

of local reflection. Also, the detectable individual

microparticle with an intensity decrease of 2.9%±0.5%

suggests a high sensitivity of using OCT to identify single

microparticle on the gold mirror surface.

In the next experiments, we investigated the feasibility of

using OCT to detect the movement of microparticles through

both clear media and skin tissue in vitro. Fig. 7 shows the

typical result from the experiments of monitoring

microparticles through water. The intensity information of the

OCT signal from the gold mirror surface can represent the

state of the microparticles within the scanning region. The

intensity (averaged from two-dimensional scanning area)

detected from the gold mirror surface covered with only water

was first recorded and used as the reference value for

normalization. From Fig. 7 (a), it can be seen that during the

time range of 0–3.5 minutes, the signal has a relatively high

frequency and small amplitude of variation, which we attribute

to the Brownian motion of the microparticles in water. When a

magnetic field is applied at about 3.5 minutes into the

experiment, the microparticles start to form long chains

(similar as the phenomenon observed in [38]) and move into

the direction of the magnet, which can be observed in Fig. 7(b)

and (c). As the microparticles are pulled out from the scanning

region, shown in Fig. 7(c), (d) and (e), the intensity signal

increases over time, and remains relatively constant after

about 14 minutes. During the time of 14–17 minutes, the left

several clusters of microparticles cannot be moved by the

magnetic field and maintain stable on the gold mirror surface,

as shown in Fig. 7(f) and (g). The fluctuation of the intensity

signal during different periods in the experiment is analyzed

in Fig. 7(h). For each period, the percentage of the signal

fluctuation was calculated with the standard deviation of the

normalized intensity divided by the relevant averaged value.

From Fig. 7(h), it can be seen that the fluctuation of the signal

from 0–3.5 minutes and 14–17 minutes is much smaller than

the one during 3.5–14 minutes when the magnetic field was

just applied to the microparticles. This is because large

amounts of microparticles in clusters were driven out of and

also into the scanning region from the time of 3.5–14 minutes,

which resulted in the signal fluctuation with relatively very

high amplitude and low frequency, as shown in Fig. 7(a). It

can also be observed that the fluctuation of the signal during

14–17 minutes appears to be half smaller than the one during

0–3.5 minutes. This can be explained by the fact that during

the last three minutes, there were fewer microparticles

presented in the scanning region, which indicates the overall

Brownian motion of the microparticles during this period is

much less than the one during the time of 0–3.5 minutes.

These results indicate that the state of microparticles on the

gold mirror surface can be clearly monitored over time

Fig. 6. (a) A plot of intensity change vs. different number of microparticles in

aggregates. For single microparticles and aggregates with 1-8 microparticles,

N=4 (samples for measurement). For aggregates with 9, 10 and 11

microparticles, N=1 (sample for measurement). (b) Optical microscope

images of microparticle aggregates used for quantifying the fractional

decrease of the intensity of OCT signal.

Fig. 7. (a) Normalized intensity profile over time indicating the movement of microparticles on the gold mirror surface in clear media (signal normalized to the

averaged intensity value with water alone). (b)–(g) En face OCT images selected from different time points. (h) The fluctuation of intensity during four periods:

reference state (water alone), 0–3.5 min (before applying magnetic field), 3.5–14 min (microparticles moving in the magnetic field), and 14–17 min (clusters of

microparticles left stable in the magnetic field).

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

through clear media based on the intensity information of the

OCT signal.

The result from the experiment of detecting the movement

of microparticles under skin tissue in vitro is presented in Fig.

8. The averaged intensity value from each two-dimensional

scan is plotted over time to represent the state of

microparticles under tissue before and after applying the

magnetic field. Similar to the result presented in Fig. 7, during

0–2 minutes (before the magnetic field was applied), the

variation of the intensity signal appears to have lower

amplitude and higher frequency relative to the one from the

time of 2–4 minutes when the magnetic field was applied, as

shown in Fig. 8(a). This is caused by the movement of

microparticle clusters over the scanning region. Also, similarly,

it can be observed that the averaged intensity increases to a

higher value after the magnetic field was applied, due to fewer

microparticles remaining within the scanning region. These

are also indicated in Fig. 8(b) where the intensity signals are

compared before and after applying the magnetic field. The

percentage of the signal fluctuation has been calculated to be

0.6% and 1.0% for the time periods before and after the

magnetic field was applied, respectively. This is comparable

with the data presented in Fig. 7(h). The differences (0.6% to

0.4% and 1.0% to 1.5%) can be resulted from the variation of

the light scattering from skin tissue over the time of

monitoring. The result presented in Fig. 8 demonstrates that

OCT can be utilized to detect the movement of microparticles

over time on the gold mirror surface under skin tissue in vitro.

These results from the feasibility study belong to the first

phase in our goal of developing optical implantable sensing

method to continuously monitor the blood sugar levels in the

subcutaneous tissue. The disadvantage of using gold mirror

lies in the challenges to position the gold mirror surface

perpendicular to the laser beam as well as to maintain exactly

the same scanning region during multiple times of

measurements. Regarding this problem, a micro-retroreflector-

based gold platform with a fixed distribution pattern has been

designed and manufactured to provide highly-reflected light

and also the possibility of implantation [33]. For the imaging

with OCT, the platform has been demonstrated to be able to

offer well-specified region and good signal-to-noise ratio [35].

The proposed glucose sensing method consists of two

information-transducing steps. The first one is from the

glucose concentration to the number of microparticles through

a competitive displacement process, and the second one is

from the microparticle number to the intensity of OCT signal.

The presented results focus on the latter step and have

demonstrated the feasibility of using OCT to detect and

monitor microparticles on the gold mirror surface both in clear

media and under skin tissue in vitro. Based on our recent

preliminary experiments using micro-channels, we have also

observed that the dextran-coated microparticles with a

diameter of ~3 m bind efficiently onto the gold surface that is

functionalized with Con A. Additionally, the presence of

glucose results in a decrease in the number of bound

microparticles on the gold surface due to a reduction in the

number of available binding sites on the Con A-functionalized

surface. In the future, our work will focus on optimizing the

chemistry for sensing glucose. We also plan to integrate our

approach with micro-retroreflector-based sensing elements

and to investigate the relationship between the OCT intensity

signal and the glucose concentration.

IV. CONCLUSIONS

We describe the concept of a new optical glucose sensing

method for continuous monitoring blood sugar levels in the

upper dermis layer. The method is based on the competitive

binding of glucose and microparticles-labelled dextran to Con

A immobilized on the surface of a highly-reflective

implantable gold platform. With the reversible binding process,

an increase in glucose concentration reduces the amount of

microparticles within the target region on the gold mirror

surface. Due to the light scattering property of microparticles,

the information of microparticle number can be revealed by

detecting the change of light reflection from the gold mirror

surface using OCT. The diameter of the microparticles and the

Fig. 8. (a) Intensity profile over time indicating the movement of microparticles on the gold mirror surface under skin tissue in vitro (intensity averaged from the

OCT two-dimensional scanning area). (b) Comparison of the intensity values before (0–2 min) and after (2–4 min) applying the magnetic field.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

central wavelength of OCT system are optimized for the

sensitivity of the detection. The feasibility of using OCT to

detect and monitor the microparticles on the gold mirror

surface has been demonstrated both in clear media and

through skin tissue in vitro. The detectable intensity change of

2.9%±0.5% caused by an individual microparticle suggests a

high sensitivity of microparticle detection using OCT. With

further development, the proposed method can potentially

offer improved sensitivity and convenience for the everyday

blood glucose monitoring.

ACKNOWLEDGMENT

This study was supported by a grant from the National

Science Foundation (CMMI-0900743) and by a grant from the

Welch Foundation (E-1264). Also, this work was partially

supported by Grant No. U54 AI057156 from NIAID/NIH. Its

contents are solely the responsibility of the authors and do not

necessarily represent the official views of the RCE Programs

Office, NIAID, or NIH.

REFERENCES

[1] A. D. Association, “Standards of Medical Care in Diabetes—2011,”

Diabetes Care, vol. 34, no. Supplement 1, pp. S11-S61, 2011.

[2] M. R. Burge, S. Mitchell, A. Sawyer et al., “Continuous Glucose

Monitoring: The Future of Diabetes Management,” Diabetes Spectrum,

vol. 21, no. 2, pp. 112-119, 2008.

[3] G. S. Wilson, and R. Gifford, “Biosensors for real-time in vivo

measurements,” Biosensors and Bioelectronics, vol. 20, no. 12, pp.

2388-2403, 2005.

[4] J. Wang, “Electrochemical Glucose Biosensors,” Chemical Reviews,

vol. 108, no. 2, pp. 814-825, 2007.

[5] M.-S. Steiner, A. Duerkop, and O. S. Wolfbeis, “Optical methods for

sensing glucose,” Chemical Society Reviews, vol. 40, no. 9, pp. 4805-

4839, 2011.

[6] R. Mc ichols, and G. Cot , “Optical glucose sensing in biological

fluids: an overview,” Journal of Biomedical Optics, vol. 5, no. 1, pp. 5-

16, 2000.

[7] A. Tura, A. Maran, and G. Pacini, “ on-invasive glucose monitoring:

Assessment of technologies and devices according to quantitative

criteria,” Diabetes Research and Clinical Practice, vol. 77, no. 1, pp.

16-40, 2007.

[8] O. S. Khalil, “ on-invasive glucose measurement technologies: an

update from 1999 to the dawn of the new millennium,” Diabetes

Technol Ther, vol. 6, no. 5, pp. 660-97, 2004.

[9] S. F. Malin, T. L. Ruchti, T. B. Blank et al., “ oninvasive Prediction

of Glucose by Near-Infrared Diffuse Reflectance Spectroscopy,”

Clinical Chemistry, vol. 45, no. 9, pp. 1651-1658, 1999.

[10] A. K. Amerov, J. Chen, G. W. Small et al., “Scattering and Absorption

Effects in the Determination of Glucose in Whole Blood by Near-

Infrared Spectroscopy,” Analytical Chemistry, vol. 77, no. 14, pp.

4587-4594, 2005.

[11] C. R. Yonzon, C. L. Haynes, X. Zhang et al., “A Glucose Biosensor

Based on Surface-Enhanced Raman Scattering:  Improved Partition

Layer, Temporal Stability, Reversibility, and Resistance to Serum

Protein Interference,” Analytical Chemistry, vol. 76, no. 1, pp. 78-85,

2003.

[12] E. B. Hanlon, R. Manoharan, T. W. Koo et al., “Prospects for in vivo

Raman spectroscopy,” Phys Med Biol, vol. 45, no. 2, pp. R1-59, 2000.

[13] B. Rabinovitch, W. F. March, and R. L. Adams, “ oninvasive Glucose

Monitoring of the Aqueous Humor of the Eye: Part I. Measurement of

Very Small Optical Rotations,” Diabetes Care, vol. 5, no. 3, pp. 254-

258, 1982.

[14] R. Rawer, W. Stork, and C. F. Kreiner, “ on-invasive polarimetric

measurement of glucose concentration in the anterior chamber of the

eye,” Graefe's Archive for Clinical and Experimental Ophthalmology,

vol. 242, no. 12, pp. 1017-1023, 2004.

[15] K. V. Larin, M. S. Eledrisi, M. Motamedi et al., “ oninvasive Blood

Glucose Monitoring With Optical Coherence Tomography,” Diabetes

Care, vol. 25, no. 12, pp. 2263-2267, 2002.

[16] R. O. Esenaliev, K. V. Larin, I. V. Larina et al., “ oninvasive

monitoring of glucose concentration with optical coherence

tomography,” Opt. Lett., vol. 26, no. 13, pp. 992-994, 2001.

[17] A. I. Kholodnykh, I. Y. Petrova, K. V. Larin et al., “Precision of

Measurement of Tissue Optical Properties with Optical Coherence

Tomography,” Appl. Opt., vol. 42, no. 16, pp. 3027-3037, 2003.

[18] K. V. Larin, M. Motamedi, T. V. Ashitkov et al., “Specificity of

noninvasive blood glucose sensing using optical coherence

tomography technique: a pilot study,” Phys Med Biol, vol. 48, no. 10,

pp. 1371-90, 2003.

[19] J. F. Sierra, J. Galbán, and J. R. Castillo, “Determination of Glucose in

Blood Based on the Intrinsic Fluorescence of Glucose Oxidase,”

Analytical Chemistry, vol. 69, no. 8, pp. 1471-1476, 1997.

[20] P. De Luca, M. Lepore, M. Portaccio et al., “Glucose Determination by

Means of Steady-state and Time-course UV Fluorescence in Free or

Immobilized Glucose Oxidase,” Sensors, vol. 7, no. 11, pp. 2612-2625,

2007.

[21] T. D. James, K. R. A. S. Sandanayake, and S. Shinkai, “ ovel

photoinduced electron-transfer sensor for saccharides based on the

interaction of boronic acid and amine,” Journal of the Chemical

Society, Chemical Communications, no. 4, pp. 477-478, 1994.

[22] H. Eggert, J. Frederiksen, C. Morin et al., “A ew Glucose-Selective

Fluorescent Bisboronic Acid. First Report of Strong α-Furanose

Complexation in Aqueous Solution at Physiological pH1,” The Journal

of Organic Chemistry, vol. 64, no. 11, pp. 3846-3852, 1999.

[23] J. S. Schultz, Optical sensor of plasma constituents, United States

Patent 4344438, 1982.

[24] J. S. Schultz, S. Mansouri, and I. J. Goldstein, “Affinity Sensor: A ew

Technique for Developing Implantable Sensors for Glucose and Other

Metabolites,” Diabetes Care, vol. 5, no. 3, pp. 245-253, 1982.

[25] S. Mansouri, and J. S. Schultz, “A Miniature Optical Glucose Sensor

Based on Affinity Binding,” Nat Biotech, vol. 2, no. 10, pp. 885-890,

1984.

[26] Champdor, Marcella, M. Staiano et al., “Thermostable Proteins as

Probe for the Design of Advanced Fluorescence Biosensors,” Reviews

in Environmental Science and Biotechnology, vol. 5, no. 2-3, pp. 233-

242, 2006.

[27] A. Sakaguchi-Mikami, A. Taneoka, R. Yamoto et al., “Engineering of

ligand specificity of periplasmic binding protein for glucose sensing,”

Biotechnology Letters, vol. 30, no. 8, pp. 1453-1460, 2008.

[28] R. Ballerstadt, C. Evans, R. McNichols et al., “Concanavalin A for in

vivo glucose sensing: A biotoxicity review,” Biosensors and

Bioelectronics, vol. 22, no. 2, pp. 275-284, 2006.

[29] R. Ballerstadt, and J. S. Schultz, “A fluorescence affinity hollow fiber

sensor for continuous transdermal glucose monitoring,” Anal Chem,

vol. 72, no. 17, pp. 4185-92, 2000.

[30] R. Ballerstadt, A. Polak, A. Beuhler et al., “In vitro long-term

performance study of a near-infrared fluorescence affinity sensor for

glucose monitoring,” Biosens Bioelectron, vol. 19, no. 8, pp. 905-14,

2004.

[31] R. Ballerstadt, C. Evans, A. Gowda et al., “In vivo performance

evaluation of a transdermal near- infrared fluorescence resonance

energy transfer affinity sensor for continuous glucose monitoring,”

Diabetes Technol Ther, vol. 8, no. 3, pp. 296-311, 2006.

[32] R. Ballerstadt, A. Kholodnykh, C. Evans et al., “Affinity-Based

Turbidity Sensor for Glucose Monitoring by Optical Coherence

Tomography:  Toward the Development of an Implantable Sensor,”

Analytical Chemistry, vol. 79, no. 18, pp. 6965-6974, 2007.

[33] T. Sherlock, A. Nasrullah, J. Litvinov et al., “Suspended, micron-scale

corner cube retroreflectors as ultra-bright optical labels,” Journal of

Vacuum Science & Technology B: Microelectronics and Nanometer

Structures, vol. 29, no. 6, pp. 06FA01-06FA01-5, 2011.

[34] J. E. Gestwicki, C. W. Cairo, L. E. Strong et al., “Influencing

Receptor−Ligand Binding Mechanisms with Multivalent Ligand

Architecture,” Journal of the American Chemical Society, vol. 124, no.

50, pp. 14922-14933, 2002.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

[35] S. N. Ivers, S. A. Baranov, T. Sherlock et al., “Depth-resolved imaging

and detection of micro-retroreflectors within biological tissue using

Optical Coherence Tomography,” Biomed Opt Express, vol. 1, no. 2,

pp. 367-377, 2010.

[36] V. V. Tuchin, Tissue optics: light scattering methods and instruments

for medical diagnosis: SPIE/International Society for Optical

Engineering, 2007.

[37] J. M. Schmitt, “Optical coherence tomography (OCT): a review,”

Selected Topics in Quantum Electronics, IEEE Journal of, vol. 5, no. 4,

pp. 1205-1215, 1999.

[38] S. Y. Park, H. Handa, and A. Sandhu, “Magneto-Optical Biosensing

Platform Based on Light Scattering from Self-Assembled Chains of

Functionalized Rotating Magnetic Beads,” Nano Letters, vol. 10, no. 2,

pp. 446-451, 2009.

Shang Wang is a Ph.D. student in Biomedical

Engineering at the University of Houston. He

received his Bachelor’s degree in Optoelectronic

Information Engineering (2010) from Harbin

Institute of Technology, China. His research

interest lies in biomedical optics, with the current

focus on the development of optical coherence

tomography and elastography methods for tissue

characterization and biomedical sensing.

Tim Sherlock received a B.S. in Engineering

Science from Trinity University (2006) and his

Ph.D. in Electrical Engineering from the

University of Houston (2011) working on the

fabrication of micro retroreflectors for use in

optical biosensors. His research interest is in

nanofabrication techniques, specifically novel

lithography and etching processes.

Betsy Salazar is a Ph.D. student in Biomedical

Engineering at the University of Houston. She

received her Bachelor’s degree in Biomedical

Engineering (2011) from the University of

Houston. Her research interests lie in

cardiovascular tissue engineering, with the current

focus of engineering cardiac tissues and enhancing

their performance with electrical, mechanical and

perfusion bioreactors.

Narendran Sudheendran is currently a Ph.D.

student in Biomedical Engineering at the

University of Houston. He received his M.S.

degree from Electrical Engineering at the

University of Houston in 2010. He is primarily

interested in designing algorithms and software

tools to study the morphological features

especially that of the cardiovascular system during

the development of mouse embryo imaged using optical coherence

tomography.

Ravi Kiran Manapuram is currently a

Development Engineer in the department of

Research and Development at Bioptigen Inc.

Manapuram received his M.S. in Electrical

Engineering (2008) and Ph.D. in Mechanical

Engineering (2012) from the University of

Houston. Manapuram’s research contributions are

in developing systems for noninvasive and nondestructive imaging and

diagnostics of tissues and cells. He has authored 14 peer-reviewed

publications and has received numerous awards including Research

Excellence Award from SPIE, selected under top 50 optics researchers by the

Nanobiophotonics summer school organized by University of Illinois at

Urbana Champagne, Appreciation award for active participation in an

organizing Committee for ECE graduate research conference at University of

Houston. Dr. Manapuram has delivered 8 talks in various conferences.

Katerina Kourentzi is a Research Assistant

Professor of Chemical and Biomolecular

Engineering at the University of Houston. Dr.

Kourentzi received her Diploma in Chemical

Engineering (1996) from the National Technical

University in Athens, Greece and her PhD in

Chemical Engineering (2002) from the University

of Houston. Dr. Kourentzi’s research interests lie

in the interface of engineering and biotechnology.

She is interested in the development of ultra

sensitive detection labels and reporters and their application in diagnostic

assays for the early diagnosis of disease.

Paul Ruchhoeft is an Associate Professor in

Electrical and Computer Engineering at the

University of Houston, where he joined the faculty

in 2001. He received his B.S. in electrical

engineering at the University of Texas at Austin in

1995 and his M.S. and Ph.D. at the University of

Houston in 1998 and 2000, respectively. Most of

his research lies in the areas of nanolithography

and nanofabrication, where he develops high-

resolution, parallel printing processes using

helium ions and atoms and the associated

processes. His current interests lie in using these patterning techniques to

develop complex systems. He is an author of more than 35 publications, holds

eight patents, and has presented his findings through more than 60 research

presentations.

Richard C. Willson is John and Rebecca Moores

Professor of Chemical & Biomolecular

Engineering, Biology & Biochemistry, and

Biomedical Engineering at the University of

Houston. He also holds joint appointments in The

Methodist Hospital Research Institute and in the

SCBMB program at Baylor College of Medicine,

and serves as Diagnostics Theme leader of the

NIH Western Regional Center of Excellence.

Willson received his B.S. (honors) and M.S. in

Chemical Engineering at Caltech, and did his Ph.D. in Chemical Engineering

and postdoctoral studies in Biology at MIT. His research interests are in

molecular recognition and its applications in diagnostics and bioseparations.

Dr. Willson is the recipient of the NSF PYI and NIH FIRST awards, and a

Fellow of the AIMBE and AAAS.

Kirill V. Larin is an Associate Professor of

Biomedical Engineering at the University of

Houston. He also holds joint appointments at the

Department of Physiology and Biophysics at

Baylor College of Medicine and Department of

Optics and Biophysics at the Saratov State

University (Russia). Larin received his first M.S.

in Laser Physics and Mathematics from the

Saratov State University (1995), his second M.S.

in Cellular Physiology and Molecular Biophysics (2001) and Ph.D. in

Biomedical Engineering (2002) from the University of Texas Medical Branch

in Galveston. Dr. Larin’s research contributions are in Biomedical Optics and

Biophotonics and development and application of various optical methods for

noninvasive and nondestructive imaging and diagnostics of tissues and cells.

He has authored more than 70 peer-reviewed publications and chapters in six

books on Biomedical Optics. Dr. Larin has received numerous awards

including Presidential Award from Russian President Boris Yeltsin, Wallace

Coulter Young Investigator Translation Award, Office of Naval Research

Young Investigator Award, Outstanding Young Investigator Award from the

Houston Society for Engineers in Medicine and Biology, and Herbert Allen

Award from American Society for Mechanical Engineers. Dr. Larin has

delivered more than 50 invited and plenary talks, serves as a chair of

Dynamics and Fluctuations in Biomedical Photonics and Optical Elastography

conferences and a member of number technical committees at professional

conferences. He is also an Instructor for short courses on Tissue Optics and

Biophotonics for the SPIE, IEEE, and OSA.